FtsZ, a homolog of tubulin, is the major cytoskeletal protein of bacterial cell division. FtsZ forms a ring around the center of the bacterium, which remains in place for most of the cell cycle. Ultimately the Z-ring constricts to divide the cell. In vitro, FtsZ assembles into long, straight protofilaments (pfs) that can associate into pf pairs and sheets. We have recently proposed a model in which FtsZ pfs assemble isodesmically, which is very different from the cooperative assembly of actin and microtubules. This model predicts a rapid fragmentation and annealing of pfs coupled to GTP hydrolysis. Using fluorescence recovery after photobleaching (FRAP) for in vivo analysis, we have also determined that FtsZ in the Z-ring is turning over rapidly, with a 30 sec halftime. This is consistent with our expectations from the isodesmic assembly model. However, that model at present is based largely on predictions from pf length, and it is essential to obtain experimental data confirming or modifying it. Most important is a direct measure of the interaction affinity of FtsZ subunits in the pf. Is it on the order of nM as predicted by the theory, or on the order of uM as predicted by other observations? We propose to use fluorescence anisotropy to measure the association of labeled subunits to pfs and estimate the Kd. We will then use the Biacore for more quantitative analysis, to determine the Kd for pf assembly, and hopefully the length of pfs assembled at different FtsZ concentrations. For several studies we will produce cap mutants that are blocked for assembly at one or the other ends. Their association into heterodimers should be a much simpler reaction than assembly of full pfs. The cap mutants will also be used to study the GTPase mechanism in vitro. Complementing these in vitro studies, we will extend our FRAP study of in vivo dynamics to new FtsZ mutants and accessory proteins. An important question that we can now address by FRAP is the state of assembly of FtsZ in the bacterial cytoplasm - is it monomers or pfs? We will determine this by diffusion measurements. Overall, we are aiming for a complete characterization of the biophysics of FtsZ protofilament assembly in vitro, and complementary analysis by FRAP of assembly dynamics and function in vivo.